OSPF Areas are used for OSPF scalability. In this post, we will look at many different topics about OSPF Area Types — their usage, limitations, and the different router types involved. In OSPF, we have in general two different area types.
OSPF Backbone Area and OSPF Non-Backbone Area. There are many different OSPF Non-Backbone Area types, and we will cover each one of them, their use case, limitations, and benefits in this post.
I strongly recommend checking the OSPF Course for those who want to learn more about this topic.
OSPF Area 0 - OSPF Backbone Area
First of all, let's start with Area 0. It is known as the Backbone area in OSPF, and if there are many different areas, non-backbone areas can communicate with each other through OSPF Area 0. Let's use the below topology for the rest of the OSPF Area Types discussion.
Figure - OSPF Area Types
In the above topology, OSPF Area 0, which is the OSPF Backbone Area, physically connects different OSPF Non-Backbone Areas. OSPF Router Types in OSPF Area 0 are known as Internal Backbone Routers. We don't have OSPF LSA restrictions when it comes to OSPF Area 0 — all the LSAs are allowed in this OSPF Area, except Type 4 and Type 7. Type 4 LSA is used for ASBR reachability in Non-Backbone areas in Multi Area OSPF, which will be explained later in the post. Before we continue with OSPF non-backbone Area Types, let's briefly mention the different OSPF Router types.
OSPF ABR
An OSPF Router which connects different types of Areas is known as OSPF ABR, OSPF Area Border Router. Based on the standard OSPF RFC, if a router has an interface in two different OSPF Areas, it is considered an OSPF ABR. Two different OSPF Areas can even be two different OSPF non-backbone Areas. But implementation on vendor routers requires a router to have at least one interface in OSPF Area 0 in order to be considered an OSPF ABR. OSPF ABR allows summarization between the Areas. Meaning, if there are many /24 subnets, for example, all of them can be summarized as /18, /16, etc., and sent as fewer prefixes to the other OSPF Area.
Sending a summary route is good for OSPF fast convergence, OSPF troubleshooting, and overall OSPF high availability. Also, OSPF ABR doesn't allow topology information to be shared between the OSPF Areas. Meaning, the OSPF Cost information of one OSPF Area is not known by other internal OSPF Area routers, because the ABR in OSPF doesn't allow this information to be shared. This is also important for the scalability of OSPF networks.
In the above topology, R3 and R4 are the OSPF ABR Routers, connecting Area 1 and Area 2 to the Backbone OSPF Area respectively. The Area ID in OSPF is important in OSPF Design; as we will see later in the post, we can create an OSPF Hierarchical design by assigning OSPF Area IDs/numbers carefully.
OSPF ASBR
OSPF ASBR, Autonomous System Boundary Router, is not actually a special OSPF Router Type. ASBR is a generic term that is used in many different architectures and frameworks in Networking. We use this term whenever one routing protocol advertises information into another routing protocol.
OSPF redistributes to EIGRP or vice versa, BGP redistributes to IS-IS or vice versa — any routing protocol redistribution is basically done on the ASBR Router. Basically, a router which connects different administrative domains is known as ASBR. When it is used in OSPF, it is called OSPF ASBR. In the above topology, R2 is the OSPF ASBR, connecting different routing domains to the OSPF network.
OSPF Stub Area
This is one of the OSPF Non-Backbone Area types. A Stub Area doesn't allow OSPF Type 5 LSA. When there is no Type 5 LSA, no Type 4 LSA is generated by the OSPF ABR into the OSPF Stub Area.
The overall reason why we use an OSPF Stub Area is that we don't want to allow redistributed prefixes in a Stub Area. But still, Type 3 LSA is allowed in an OSPF Stub Area.
OSPF Totally Stubby Area
Another Non-Backbone area in OSPF is the OSPF Totally Stubby Area. In the Totally Stubby area, in addition to Type 5 and Type 4 LSA, Type 3 LSAs are not allowed either. So, in addition to the redistributed prefixes, Inter-Area OSPF prefixes are not allowed either.
Why don't we allow Inter-Area prefixes in a Totally Stubby Area in OSPF?
The reason is again scalability. Whenever you restrict more LSAs, it is good for overall network scalability. It might create sub-optimal routing, but network design is all about trade-offs. We lose something, we gain another. Let's have a look at Single Area and Multi-Area OSPF next.
Single Area OSPF
We can place all our routers in a Single Area OSPF. It is also commonly referred to as a Flat OSPF Network Design — all routers in OSPF Area 0, or in any other single OSPF Area. This means that if you place all your routers in OSPF Area 100, it is still considered a Single Area OSPF, because each and every router inside that OSPF Area will share the same OSPF database, the same reachability information — basically the same everything about each other.
OSPF Single Area / Flat design may not be scalable after a certain number of OSPF Routers and prefixes. Thus, for scalability, we usually consider hierarchy. A common way to deploy hierarchy in OSPF is explained next.
Multi Area OSPF
OSPF Multi-Area design involves OSPF Area 0, the Backbone OSPF Area, as well as Non-Backbone Areas. In this design, topology information is not exchanged. The cost between the routers in OSPF Area 0 is only known by Internal OSPF Area 0 routers, not by any other type of OSPF Area in the network.
Also, Network Admins usually summarize the prefixes between the OSPF Areas, so less information is exchanged between the areas. Or they deploy the Non-backbone OSPF Areas as Totally Stubby Areas or NSSA areas, so they just send the default route into those OSPF Areas. Multi-Area OSPF design is covered here in more detail.
OSPF NSSA Area
The OSPF NSSA Area is also a Stub area, but it is called an OSPF Not So Stubby Area. You will understand the reason below. When we don't want to allow Type 5 LSA — which means redistribution from the Backbone Area — but still want to allow redistribution into the Non-Backbone OSPF Area, then we deploy an OSPF NSSA Area. NSSA OSPF allows redistribution from different domains, but redistributed prefixes can't come from the Network Backbone / OSPF Area 0.
A router that connects an OSPF NSSA Area and OSPF Area 0 is known as an OSPF NSSA ABR. Redistributed prefixes in NSSA are marked as Type 7 LSA. The OSPF NSSA ABR translates OSPF Type 7 LSA to OSPF Type 5 LSA. If there is more than one OSPF NSSA ABR, the Router with the lower OSPF Router ID translates from Type 7 to Type 5.
Why Use OSPF Stub Areas?
Restricting the LSA types allowed into an area is not just an academic exercise — stub areas deliver concrete benefits, particularly in large and complex networks. The first and foremost advantage is the reduction in resource consumption. By configuring some areas as stub, the number of route entries managed by each router decreases, easing memory and processing requirements. Reducing the frequency and scope of routing updates also minimizes the chance of routing loops.
Stub areas also improve network stability and predictability. With fewer routes advertised across the OSPF network, there is less fluctuation, leading to a more stable environment — which is key for critical applications that rely on consistent performance. Fewer routes to process also means faster convergence, since routers can reach a stable state more quickly after a change or failure. Finally, a simpler routing table makes the area easier to manage and troubleshoot.
When to Use Stub, Totally Stubby, or NSSA Areas
Not every network benefits from a stub configuration, and understanding when to use each type is what makes the difference. Smaller networks with a simple topology may see little advantage, as the overhead saved is minimal. In large enterprise environments with multiple branch offices, or in service provider networks where scaling is paramount, stub areas can significantly optimize performance.
Consider a corporate headquarters with robust computing resources connected to various branch offices running less powerful hardware. By configuring the branch office connections as stub or totally stubby areas, you ensure that these sites are not burdened with heavy routing tables and complex OSPF calculations they might not fully use. A totally stubby area is well suited to remote locations that only need a default route to reach the rest of the network. Where a local site must advertise its own external routes — for example, routes redistributed from another protocol at a branch — an NSSA is the better choice, because it permits external routes as Type 7 LSAs while still keeping the area lean.
Configuring OSPF Stub Areas
Although the exact commands vary slightly depending on the hardware and software, the general approach to configuring a stub area is fairly standard:
- Define the area as a stub: this is done on all routers within the stub area, marking it accordingly in the OSPF configuration.
- Limit the propagation of specific LSAs: configure which link-state advertisement types are allowed into and out of the area, depending on whether it is a stub, totally stubby, or NSSA.
- Verify and monitor: after the change, always verify with OSPF show commands and monitor operation to ensure everything works as expected.
Because misconfiguration can lead to communication failures or sub-optimal routing, regular monitoring and troubleshooting are essential. Testing the design in a lab or with network simulation tools beforehand helps anticipate operational issues before they reach the live network.
OSPF Network Types
Area types describe how routing information is scoped across an OSPF domain; OSPF network types, by contrast, describe how OSPF behaves on a given link or segment. Understanding both is useful when designing an OSPF network.
Broadcast
The Broadcast network type is typically used in LAN environments where communication is multi-access. Routers use a multicast address to exchange OSPF information, and a Designated Router (DR) and Backup Designated Router (BDR) are elected to reduce the number of OSPF updates and decrease network overhead.
Non-Broadcast
The Non-Broadcast network type does not assume that all routers can reach each other directly. It is common in WAN environments such as Frame Relay or ATM, where multicast or broadcast is not inherently supported. Non-Broadcast networks require manual configuration of neighbors — more administrative work, but better control over network resources and OSPF traffic.
Point-to-Multipoint
Point-to-Multipoint treats each connection as a separate point-to-point link. This is particularly useful in partially meshed networks such as hub-and-spoke or satellite topologies. It simplifies OSPF management by removing the need for a DR or BDR.
Point-to-Point
The Point-to-Point network type is used for direct router-to-router connections, such as leased lines. There is no election of DR and BDR, which minimizes network traffic and simplifies troubleshooting — ideal where control and predictability are paramount.
Conclusion
OSPF areas and area types are ultimately about scalability. The Backbone Area ties everything together, ABRs and ASBRs define the boundaries, and stub, totally stubby, and NSSA areas let you trade route detail for a leaner, more stable, faster-converging network. Choosing the right area type — and the right network type on each link — should always be aligned with the broader design goals of the network. As the network grows, continuously reassessing these choices is what keeps OSPF efficient and manageable.
